Electric Water pumps

[Graham Perry]

Has anyone any experience of electric water pumps on their cars.

My crossflow has always been a difficult car to keep cool even though I am using all the usual tricks to do so (Ali Rad,larger electric fan with override,headertank, water wetter, larger heater radiator etc) and I find that there always seems to be a sprint on the hottest day of the year so that even the above are not sufficient.

 

I was thinking of fitting one so that when I am sitting in a queue waiting for a run I can have it going flat out so that the engine isn't too hot before I even start.

 

Are they worth it and roughly how much power do they save ?

[Mike Bees]

Dave Walker tested one on their Fiesta Stock Hatch a while back. I think he did publish the results in CCC. Sadly I can't remember whether the answer was 'bugger all' or '3 bhp'.

 

Mike

[Roger King]

Graham,

 

There should be no problem with cooling on a Crossflow. We have customers out with steel 185bhp engines who are doing the Jaguar Car Club Centurion (100 mile) races. Even on extremely hot days (not this year obviously!), cooling is OK.

 

They are not running anything fancy by way of cooling. In the main they use the Caterham competition radiator with a sealed header tank system. The only special mod is to move the oil cooler above and behind the water radiator. This reduces the oil cooler's efficiency quite a lot, but with the better water cooling it is still adequate with the right oil.

 

I suspect that you have some simple problem that could easily be cured (maybe the gauge is reading incorrectly).

 

If you e-mail me on rkingeng@aol.com with details of running temperature, thermostat type, etc, I may be able to advise.

 

 

Roger King.

[Red SLR]

I have seen these pumps around. I think they claim "upto" 10% gain.

 

I have not seen one on a Caterham yet but I dont like the build quality.

 

Simon.

 

X777CAT

[Peter Carmichael]

I am very suspicious of the electric coolant pump system that has cropped up recently. Here's why:

 

The coolant system transports heat away from the engine. The coolant/engine surface interfaces should give a very good heat transfer. Transporting the heat out of the engine should be the easy bit. That heat stays in the coolant unless you can get the radiator to dissipate it. If you increase the flow rate, you don't necessarily improve the heat transfer in the water/air heat exchanger (radiator - Tony Martyr pointed out what a stupid name that was for it).

 

If the coolant flow rate is higher, for any given heat transfer the temperature difference across the block will be less. The heat transfer across the radiator has to match the heat transfer in the block and the temperature difference is the same in reverse. If you are trying to reduce the maximum temperature in the system, then your radiator is not going to be working as well - the radiator relies on a temperature difference to ambient air temp.

 

Managing the variation in heat dissipation that is required between idling and WOT at max rpm is complex and airflow through the radiator typically rises with rising power. i.e. all cars would overheat in seconds if sinking maximum power at standstill. A thermostat is a clever component in a closed loop feedback system - configuring an electric pump in any open loop mode is a bad idea.

 

However, crossflows don't generally overheat, just like most other engines don't overheat. If all the components in the cooling system are working correctly then there must be too much heat getting into the system, even at idle. This suggests that something pretty screwy is going on with your thermal efficiency at idle.

 

That was a big 'if' that I put in back there and it is difficult to assess every component, especially when you think you have tried everything already. It is difficult to assess flow rates. You cannot trust the Caterham temperature gauge and it only measures the temperature in one place.

 

If you can get an accurate measure of head outflow temperature, you can work out whether the thermostat is working: Start up the engine from cold and take temperature measurements every 15 seconds. Feeling for when the thermostat opens by checking the top hose to the radiator, you should notice the temperature stabilise at the thermostat's rated temperature as cold water from the radiator is bled into the head circuit. If this doesn't happen then the flow rates are absolutely screwy or the thermostat is not working.

 

If you know the volume of your radiator and you know the heat capacity of the coolant in it, you can work out an approximate figure for the heat production at idle according to the time it takes until the temperature starts to climb again because the thermostat is fully open and all the water in the radiator circuit is now up to the temperature of the rest of the circuit. (For this test to work, you don't want the fan to be running...) You could cross-check with another crossflow owner.

 

Thermal efficiency can be fundamentally affected by cam timing but you would have to be quite a long way out.

 

If it is a problem with temperature at idle, it is probably the lack of airflow that is the problem. Moving the oil cooler sounds like a shrewd move, but you would expect RK to know.

[old captain slow]

In my view you gets nothing for nothing in this world, so unless the electric pump is MUCH more efficient than the mechanical one the engine is loaded by the same amount to drive the pump either way (via the alternator for the electric pump). So unless you disconnect the alternator drive belt for racing than the gain will be around about zero. thumbsdown.gif

 

Agree?

[CHRIS CLARK]

For further info. try:

 

www.daviescraig.com.au/newproduct.html

 

Claims of up to 20HP (Yeah right!) but who knows?

I believe these systems are sold in UK by "Demon Tweeks"

 

[Graham Perry]

Dear all, Thanks for your comments its more complex than I had considered. I had naively looked at it a different way. Say in a typical engine cycle over a one minute period it fires say 10000 times at varying RPM then I thought that if you could increase the number of visits to the heat exhanger that every piece of water makes per 10000 ignitions then you would dissipate more heat. I had believed that the speed of the water through the exchanger is not as important as the number of visits each piece of water makes to the exchanger. I know of a thermodynamisist here at work. If I can collar him I will ask his views.

 

One advantage that the electric pumps do have though, apart from a negligable power gain is that they are of an efficient design and do not cavitate. Something that a typical pressed steel impeller will do at at a surprisingly low RPM.Cavitation is the last thing you want on a 9000 RPM engine.

 

My car is fine normally, its just at one particular event every year I have to queue for typcally 25 minutes and typically its always hot. At pull away the car is so hot (gauge in the red) its pinking slightly and obviously on the verge of a detonation, so I was just trying to engineer my way round the problem rather than re-map the ignition for a situation that occurs typically once a year. The engine cooling system is fine and the gauge accurate to within about 5 degrees. On every strip down its been thoroughly checked and nothing found to be blocked or cracked and no sign of lean running (Set up beautifully up by RK himself) detonation or Oil starvation. Its a problem that occurs only in extremis, but no-one wants their engine to go bang !

 

[Mike Bees]

A couple of questions spring to mind Graham. (1) Why not switch it off while you're in the queue to avoid it getting red hot(this seems too obvious, I know you're not daft!). (2) If your ECU has provision for temperature correction then it's easy to reduce the advance as the coolant temperature (or underbonnet air temperature) rises above what you consider to be the danger point.

 

Mike

[Peter Carmichael]

>>>I had believed that the speed of the water through the exchanger is not as important as the number of visits each piece of water makes to the exchanger. I know of a thermodynamisist here at work. If I can collar him I will ask his views.<<<

 

Oh lordy! Heat exchangers are b*stard things to understand and confound intuitive speculation. What you have to remember is that each fluid (coolant or air) doesn't know about the other fluid. All they know about is the circumstance of their own flow and the temperature of the surfaces that they come into contact with.

 

If you have a bulk of coolant flow which is too hot, and you want to shift heat out of it it is no good having a laminar flow through a large channel because the majority of the coolant will never come into contact with the surface. Consideration of heat exchangers starts with the calculation of Reynolds number, giving an indication of the turbulence of the flow. You then have to consider the heat transport into the fluid and you get into an area of calculation which is indeterminate - you have to guess a heat transfer power and then see, after all the calculations of Nusselt numbers and everything, whether you have actually achieved the intended transfer. This sort of analysis works reasonably well when you have a controlled liquid/liquid contraflow shell/tube heat exchanger and becomes more and more of a fudge when you have airflow at a rate that you can't really control, going across the flow of the liquid channels in the radiator.

 

Talk in the pub last night was that the high speed airflow characteristics are all wrong. What you really need is a letterbox slot in the front of the car, with an appropriately profiled duct leading to the radiator. Best cooling apparently occurs with airspeeds of about 70mph. With higher speeds you get a lot of drag with a lump of air that can't go anywhere stagnant on the front of the radiator - think about how big the grille openings are with modern car designs...

 

(This isn't answering anything, but I thought it was interesting background info.)

[Graham Perry]

Mike, Two good points. I don't tend to stop it when it gets hot because it can be a sod to start again, but having said that I should anticipate what is going on anyway and stop it before it gets too hot. I had forgotten the temp corection facility on the emerald ignition. I know its 'greyed out' currently in my software but I may enquire if it can be activated. Nice one mate.

I guess I am just a bit fussy at heart but to paraphrase Ron Dennis in his TV advert, 'we look for perfection and when we have found it we start again'

 

[anthony1956]

interesting thread. I am reading it with long distance (meaning weeks not days) self-supported touring in mind. If the water pump fails on the BDR it's replacement involves a 15 ton press, not that they don't have those outside Blighty, but not so easy. So an electric pump has its advantages from my disaster planning point of view.

 

anyway, more here* -

 

Edited by - anthonym on 14 Sep 2012 15:14:21

[Red SLR]

There are some nice alloy ones around now - build quality looks a lot better than when I last looked.. still not seen one on a 7...

[anthony1956]

interesting.. now googling.

[Surreptitious]

Just a thought - do you have a manual override switch for the cooling fan?

It's always easier to keep an engine from getting hot rather than try to bring it down from a higher temperature.

My 7 doesn't have an otter switch just a manual switch. Whenever I see a queue or am entering somewhere like Milan, I just switch the fan on.

[The Pikey]

Fitted a Stewart EWP in my BEC 7 a few years ago. It's a very well made item and worked perfectly. Since selling the car I had a gook look over a Prodrive works car and there was one fitted to that, although the did cover the name plate up with a piece of carbon

 

Jason

[The Pikey]

Stewart pump.

[Mark w]

Is this one of the longest time threads ever ?

 

Just under 12 years long at the moment

[Graham Perry]

And I can give a long term answer as I started the thread 12 years ago. I did go for an electric water pump and it did make a major difference to my crossflow cooling as the crossflow standard pump is a pretty poor design for high revs. Had it well over 10 years now and I am happy with it.

 

The thing I do like about it in particular, is that after coming off track with the car hot you can stop the engine but leave the pump running to circulate the coolant for 5 minutes or so which probably does the head gasket no harm at all.

 

I also rigged up a switch to override the electric controller so I can have it pumping all the time when on track. The controller is best left on for road use or the engine runs too cool. The controller switches the pump on or off as needed based upon the temperature in the cyclinder head.

 

Not sure its suitable for all engines as most modern engines have efficiently designed pumps anyway, but it worked for my crossflow.

[anthony1956]

what pump have you fitted Graham?

On my run today I noticed the temp rise with the rate of progress so wondering if this might help.

[Graham Perry]

Mines the original Davies Craig one.

[anthony1956]

thanks which one?

 

http://www.daviescraig.com.au/Electric_Water_Pumps-list.aspx

 

Edited by - anthonym on 20 Sep 2012 17:14:34